The primary goal of this R01 is to improve understanding of the neurobiology and clinical utility of recognizing two distinct types of primary progressive apraxia of speech (PAOS). Phonetic PAOS is characterized predominantly by distorted sound substitutions and additions, whereas Prosodic PAOS is characterized predominantly by slow, prosodically segmented speech (previously referred to as type 1 and 2, respectively; a third type is characterized by a relatively equal combination of the Phonetic and Prosodic characteristics). Little is known about these PAOS types; however, pilot data suggest biological and clinically meaningful differences between PAOS types. Specifically, Phonetic PAOS seems to be related to degeneration of neocortex, while Prosodic PAOS appears to be more subcortically and brainstem mediated. Pathological underpinnings may also differ across PAOS types. There is some evidence that Prosodic, and not Phonetic, PAOS is associated with the development of a devastating extrapyramidal syndrome and shortened survival. Our approach to the understanding of PAOS types will involve a comprehensive longitudinal assessment of clinical, neuroanatomical, functional, molecular and histopathological data for these patients. By the end of the R01, we expect to have collected and analyzed clinical data - including demographic, speech and language (perceptual and acoustic), neurological, and neuropsychological variables - for 80 PAOS patients. Of these 80, 33 have already been recruited and 47 will be recruited via this R01 mechanism. All 47 new patients will complete the identical volumetric brain MRI protocol which will allow us to assess grey matter atrophy on structural MRI, white matter tract degeneration on diffusion tensor imaging, and functional network disruption on task free fMRI. All new patients to be recruited via this R01 mechanism will also complete a dopamine transporter SPECT scan to assess for striatal dopamine receptor integrity. All tests will be completed annually. Postmortem brain examinations and additional histological analyses of specific brain regions will also be performed on the PAOS patients who are expected to die during this R01. This will be the first study to systematically investigate PAOS types, as well as follow the course of disease longitudinally, and hence is highly novel. The PI of this grant, Dr. Josephs, will be working with a team of experts in AOS (Drs. Duffy and Utianski), structural neuroimaging (Dr. Whitwell), functional neuroimaging (Dr. Jones), molecular neuroimaging (Dr. Lowe), neuropsychology (Drs. Machulda and Butts), and neuropathology (Dr. Dickson) who will work among state of the art facilities and equipment to collectively to reach the aims. At the completion of the R01, we will 1) better understand the neurobiology of PAOS and 2) validate the clinical validity and utility of PAOS types through perceptual consensus, acoustic correlates, and data driven analysis to support prognostication and the development of targeted treatments.